You have to be so careful with track laying.

[goldfish]

1 hour ago, Robert Stokes said:

 

Could you please explain how you got this figure as it seems very high. It would imply an expansion of 2mm for a yard of track if you laid it at 10 C and the temperature went up to 30 C.

    

Nickel silver is actually 60% copper, 20% nickel, 20% zinc. The expansion rates for these metals are 0.0000165, 0.0000134 and 0.0000302 cm per cm per degree Celsius. Allowing for the composition this averages at about 0.000019. This gives for a yard of track (call it 90 cm) over the same 10 to 30 C range

   0.0000193 x 90 x 20 = 0.034 cm 0r 034 mm.

so enough to matter but no where near as much as you figure.

    

 

I think you you may have the decimal point in the wrong place, it isn't shown.

 

The Linear Thermal Expansion Coefficient of Nickel Silver is 0.0000162 - 0.0000167.

https://psec.uchicago.edu/thermal_coefficients/cte_metals_05517-90143.pdf

 

There is useful calculator here :  https://www.omnicalculator.com/physics/thermal-expansion

 

The actual expansion is 0.2916mm - 0.3006mm. More than enough to matter.

[Robert Stokes]

There was a decimal point missing in my mm result (which I have now corrected) but not in the cm figure. My final answer of 0.34 mm is not far different from yours assuming that you are using the same 20C temperature range. As I said, enough to matter, but well below the 2mm implied by a previous answer.

[The Johnster]

Having had an attic layout destroyed by excessive temperature range back in the 60s, I am all to aware of the problem.  Since then, I have made a point of only ever having layouts situated in the living area of the house/flat, which is heated and ventilated,  so that the overall temperature range is lessened.  I realise that not everyone can do this, and that it may result in a compromise where the layout has to be 'put away' after operating sessions, which leads to a whole nother set of issues, but there are significant advantatges.

 

The temperature range and controlled humidity are obvious ones, but the proximity to the rest of the household, and, if the layout is a permanent erection, it's instant availabilyt for operation or work lead to a less 'formal' and more relaxed approach, and a better integration of your hobby into your life.  You can watch tv with the missis and pop out to complete a quick shunting move during the 'verts, and remain in permanent contact, with the result that your partner is less likely to resent the time you spend alone in the railway room.  My Squeeze is very supportive of my model railway activity, but less so of the time I 'waste' on this website not listening properly to her or doing my chores.

 

What chores?

 

Mine's a double scotch, thank you very much, two ice cubes no water.

[Andymsa]

The best solution is air conditioning, not everyone can afford to install it or for practical reasons Install it. My point is I have air con and a few years ago it developed a fault unfortunately during a very very hot period, but because I had left enough space for expansion on the irjs no issues were encountered. As a side note air con is very cost effectI’ve for heating or cooling. 

Edited October 29, 2021 by Andymsa

[davknigh]

While it is useful to know the linear thermal expansion coefficient of various rail materials, the culprit in most track issues is not the rail but the bench work which is more likely expand or contract especially if the humidity of the railway room is not controlled.

 

Cheers,

 

David

[BR60103]

Taking the figures above, and applying Pythagoras. (and approximating a yard to a metre)

if x is the sideways movement and the cold rail is 1000 mm; the warm rail is 1000.34mm,

1000.34 squared = 1000 squared plus x squared

x=sqrt(1000.34**2 - 1000 **2) = sqrt (680.1156) = 26.08

Which means that the rail would shift over an inch.

 

Where's my faulty reasoning?

 

 

[AyJay]

Good morning all,

I have found this a most interesting topic, especially the different ideas posted on track, wheels, couplings, as I also have 'mysterious de-railings'.

Most frustrating are those where it is just one wagon/bogie in a rake and I can see nothing obvious in the offending item.

I do find the occasional dog leg.  When I laid my track, I anticipated this problem by planning for the tightest part of the curve to be in one continuous length of flexitrack (Peco, finescale, OO gauge, droppers on every length). Thus I could, almost always, confine joins to the gentler curves.  The exception being that my layout is composed of connecting modules, not more than 1.1m across and where it happened to be a corner, then I'm technically breaking that rule, but I think I kept my radii not too tight.  Would the gentleman who advocated offsetting the rail joins please share how he performed this technique, as I am struggling getting my head around how to do this?

 

The two biggest causes of derailments for me are:

 

1. is at the joins to the boards. The track is laid on cork and at the board edges I have pinned and glued (2-part epoxy resin) lengths of copper strip on top of glued strips of softwood that were planed after fixing, so that the copper strip was at the right height for the track to sit on top and be soldered in place.  This was fine at the time, but now a small 'hump' has appeared in the track at one place over such a join, causing the wheels of one locomotive to lift off the track in the trailing 'dip' at that place and stall; a couple of trucks derail there.  Being a lazy g1t and wanting to avoid major surgery, I have tried prising up the track on the trailing side of the hump to lessen its effect, but I don't think its lessened the problem.

 

2. Where a length of track crosses a baseboard joint at an angle i.e. not at 90 degrees.... My method was to lay a length across the join and secure in place. Then when all tracks were laid, I would go down the join with a razor saw.  Where the track is perpendicular to the join, this was just fine.  But where it crossed at an angle, this meant the the cut was not perpendicular across the rail and so left a 'lip' that could catch the flange of a wheel. If I do another layout, I shall use a dremel and cutting disk to cut each rail individually.

 

I like the idea of soldering at joins and have done so where there is a short length near a baseboard edge that connects to a point. However, There is still the need to maintain at least one expansion joint per track length per baseboard. My layout is not in a controlled environment, fortunately not exposed to direct sunlight.  When I laid the track, I left 1mm expansion gaps.

 

A most interesting discussion, thank you.

Alan

 

[goldfish]

9 hours ago, BR60103 said:

Taking the figures above, and applying Pythagoras. (and approximating a yard to a metre)

if x is the sideways movement and the cold rail is 1000 mm; the warm rail is 1000.34mm,

1000.34 squared = 1000 squared plus x squared

x=sqrt(1000.34**2 - 1000 **2) = sqrt (680.1156) = 26.08

Which means that the rail would shift over an inch.

 

Where's my faulty reasoning?

 

 

 

The rail is not expanding in isolation, everything around it is also expanding to some extent at the same time. Also if the rail is constrained and cannot freely expand the heat energy will be converted into strain energy until something gives and then the rail will distort.

 

Your use of Pythagoras to model a simplified case is for one right angle triangle, but for this simplified case you should be using two right angle triangles back to back. In practice the expanded rail would form a curve and the displacement would be less.

 

Edited October 30, 2021 by goldfish

[Il Grifone]

I appear to have got the decimal point in the wrong place, but I did say, "For what it's worth" and did give a link and quote the prototypical figure. (Obviously for steel, nickel silver would be a bit greater.) This would convert to about 1/3mm for a yard/900mm of track in 4mm scale.

Apologies....

 

Feeble excuse following - feel free to ignore! It has nothing to do with railways.

 

The brain was trying to sort the circuit diagram of my B&O 2400-2 tuner amplifier (2x A4 of tiny print, illegible (by me anyway) without a magnifying glass - luckily in English, not Danish!). It had decided to go 'plop' and die when switched on. "Just a fuse", I thought - they are rather ancient - I've had the thing twenty years and it was no fledgling when I bought it! (Like me!) Needless to say they were all OK and it decided to come on again, when I was investigating the absence of the 'switch on' voltage. SWMBO whinging about the table, means I still haven't replaced the capacitor that causes a slight mains hum....

Looking on YouTube someone suggested replacing a smoothing capacitor. The advice for these (and all other vintage B&O devices*) is to replace all the electrolytics.

* Really this good advice for anything....

[Phil Bullock]

1 hour ago, AyJay said:

Good morning all,

I have found this a most interesting topic, especially the different ideas posted on track, wheels, couplings, as I also have 'mysterious de-railings'.

Most frustrating are those where it is just one wagon/bogie in a rake and I can see nothing obvious in the offending item.

I do find the occasional dog leg.  When I laid my track, I anticipated this problem by planning for the tightest part of the curve to be in one continuous length of flexitrack (Peco, finescale, OO gauge, droppers on every length). Thus I could, almost always, confine joins to the gentler curves.  The exception being that my layout is composed of connecting modules, not more than 1.1m across and where it happened to be a corner, then I'm technically breaking that rule, but I think I kept my radii not too tight.  Would the gentleman who advocated offsetting the rail joins please share how he performed this technique, as I am struggling getting my head around how to do this?

 

The two biggest causes of derailments for me are:

 

1. is at the joins to the boards. The track is laid on cork and at the board edges I have pinned and glued (2-part epoxy resin) lengths of copper strip on top of glued strips of softwood that were planed after fixing, so that the copper strip was at the right height for the track to sit on top and be soldered in place.  This was fine at the time, but now a small 'hump' has appeared in the track at one place over such a join, causing the wheels of one locomotive to lift off the track in the trailing 'dip' at that place and stall; a couple of trucks derail there.  Being a lazy g1t and wanting to avoid major surgery, I have tried prising up the track on the trailing side of the hump to lessen its effect, but I don't think its lessened the problem.

 

2. Where a length of track crosses a baseboard joint at an angle i.e. not at 90 degrees.... My method was to lay a length across the join and secure in place. Then when all tracks were laid, I would go down the join with a razor saw.  Where the track is perpendicular to the join, this was just fine.  But where it crossed at an angle, this meant the the cut was not perpendicular across the rail and so left a 'lip' that could catch the flange of a wheel. If I do another layout, I shall use a dremel and cutting disk to cut each rail individually.

 

I like the idea of soldering at joins and have done so where there is a short length near a baseboard edge that connects to a point. However, There is still the need to maintain at least one expansion joint per track length per baseboard. My layout is not in a controlled environment, fortunately not exposed to direct sunlight.  When I laid the track, I left 1mm expansion gaps.

 

A most interesting discussion, thank you.

Alan

 

Morning Alan

 

Is it always the same wagon that derails ? If so will be something out with the wagon …. Most likely couplings or wheels - can be a real pain! 

[TheSignalEngineer]

4 hours ago, Phil Bullock said:

Morning Alan

 

Is it always the same wagon that derails ? If so will be something out with the wagon …. Most likely couplings or wheels - can be a real pain! 

I had repeated problems with on set of points. Always the same loco. Turned out to have a slight distortion to the mazak chassis casting. Later I had another loco derailing there and that also had a duff chassis.

Other problems as you say with couplings, particularly Hornby close couplings sticking one way, and wheel back to backs. Wayward pony truckd are also a pain.

[AyJay]

The loco that stalls is a short wheelbase, 4 driving wheels, meaning that as it goes over the 'ridge', it only has one pair of driving wheels in contact, so it is prone to stall.  I have a tendered loco with pony truck that tends to bounce as it goes forward, and so will tend to jump the rails at the slightest imperfection; pity Hornby did not have the foresight to add a spring to hold it down on the track.

As for the errant wagon, I'm not going to lose any sleep over it.  The bogie on one coach is another matter and will stand further examination.

 

Thinking further about soldering rains together, I have in mind something like this:

Push the sleepers back to expose a few inches of rail, using some water to reduce friction and prevent damage if needed.

Clean ends with glass fibre brush, as required.

Offer up to another track length similarly prepared.  Yet allowing for the joins on both rails not to be at the same place. Also allow for a short length of sleepers between where one rail join is an where the other is. (Difficult to explain, but hope you get what I mean.

Using fishplates that are a tight, resistance, fit, add flux and solder together.  This may also be a suitable place to add the wire dropper, to the underside of the fishplate.

Push sleepers back into place.

Put track into place, cutting the other ends to fit.

[Jol Wilkinson]

I have only used flexi-track (C&L) on fiddle yard traversers but built all the scenic trackwork using a mix of plastic chairs or rivets with ply sleepers. All trackwork was joined with cosmetic plastic fishplates where insulated joints were required and with soldered, etched, fishplates when joining rail panels. This was on an end to end layout of 34ft overall, with a 22ft scenic section. The track was laid onto 3mm thick closed foam underlay with a rubber carpet underlay glue.

 

The trackwork was usually built and laid at around 18 -22 degrees ambient, stored as low as freezing in a ventilated wooden shed and set up in a local hall (too big for my house)/displayed at "exhibition hall" temperatures (as low as about 10 degrees at Warley on a Friday evening, but generally around 20+ degrees. I never experienced any expansion/contraction problems, other than on the fiddle yard traversers where the straight flexi-track was rigidly joined along its 54" lengths and anchored at the ends onto pcb sleepers. Very occasionally a rail would become "un-soldered" at the pcb sleeper through expansion/contraction. 

 

So I would suggest that expansion/contraction shouldn't present a problem for most layouts, especially if kept at reasonably consistent temperatures.

 

The issue with curved flexi-track misalignment at joints should be overcome by by curving the track sufficiently before laying and fixing it (and the underlay) securely with a rubber based glue (e.g. Copydex, Ball's F3 Styccobond carpet underlay adhesive).

[Il Grifone]

Staggered joints require one rail to be slid through the rail fixings and probably needs one rail to be cut short. This would depend on the radius.

 

I laid my last track this summer. With temperatures in the thirties, I didn't have to worry about expansion. The tight curves each end have been laid loosely, so that any expansion/contraction just moves the curve slightly sideways.

 

Any vehicle that derails should be 'red-carded' and the fault investigated. The usual causes are skew axles (either horizontally or vertically), wheel back to back out of tolerance or insufficient weight on one or more axles. Three point suspension (or springing) is advisable, especially on long wheelbase or bogie vehicles. Tension lock couplings which pull the vehicles off centre don't help. Other types that stick have the same effect.

[BR60103]

Long time ago, I read about a club who adopted the following:

Any derailment was marked with a tag on the end of the vehicle that derailed and another tag by the track where it derailed.  Any object that accumulated multiple tags would be removed from service and examined (I assume track would just be examined).

That would give an idea which bit was more likely to be causing problems, or if one vehicle and one bit of track just didn't like each other.

 

[RobinofLoxley]

On 30/10/2021 at 09:37, Il Grifone said:

I appear to have got the decimal point in the wrong place, but I did say, "For what it's worth" and did give a link and quote the prototypical figure. (Obviously for steel, nickel silver would be a bit greater.) This would convert to about 1/3mm for a yard/900mm of track in 4mm scale.

Apologies....

 

Feeble excuse following - feel free to ignore! It has nothing to do with railways.

 

The brain was trying to sort the circuit diagram of my B&O 2400-2 tuner amplifier (2x A4 of tiny print, illegible (by me anyway) without a magnifying glass - luckily in English, not Danish!). It had decided to go 'plop' and die when switched on. "Just a fuse", I thought - they are rather ancient - I've had the thing twenty years and it was no fledgling when I bought it! (Like me!) Needless to say they were all OK and it decided to come on again, when I was investigating the absence of the 'switch on' voltage. SWMBO whinging about the table, means I still haven't replaced the capacitor that causes a slight mains hum....

Looking on YouTube someone suggested replacing a smoothing capacitor. The advice for these (and all other vintage B&O devices*) is to replace all the electrolytics.

* Really this good advice for anything....

I recapped (repleaced all the capacitors) a 30 year NAD amplifier earlier this year. Sounded like new afterwards - 'sounded' in this case meaning absence of sound.

[RobinofLoxley]

22 hours ago, Jol Wilkinson said:

I have only used flexi-track (C&L) on fiddle yard traversers but built all the scenic trackwork using a mix of plastic chairs or rivets with ply sleepers. All trackwork was joined with cosmetic plastic fishplates where insulated joints were required and with soldered, etched, fishplates when joining rail panels. This was on an end to end layout of 34ft overall, with a 22ft scenic section. The track was laid onto 3mm thick closed foam underlay with a rubber carpet underlay glue.

 

The trackwork was usually built and laid at around 18 -22 degrees ambient, stored as low as freezing in a ventilated wooden shed and set up in a local hall (too big for my house)/displayed at "exhibition hall" temperatures (as low as about 10 degrees at Warley on a Friday evening, but generally around 20+ degrees. I never experienced any expansion/contraction problems, other than on the fiddle yard traversers where the straight flexi-track was rigidly joined along its 54" lengths and anchored at the ends onto pcb sleepers. Very occasionally a rail would become "un-soldered" at the pcb sleeper through expansion/contraction. 

 

So I would suggest that expansion/contraction shouldn't present a problem for most layouts, especially if kept at reasonably consistent temperatures.

 

The issue with curved flexi-track misalignment at joints should be overcome by by curving the track sufficiently before laying and fixing it (and the underlay) securely with a rubber based glue (e.g. Copydex, Ball's F3 Styccobond carpet underlay adhesive).

Any plastic glue will creep under load to some extent. Mechanical fixing, ie track pins, may be less glamorous but its surely the most effective

[The Johnster]

If your running is good and reliable in general, without frequent derailments (and this will nearly always be the result of care taken in track laying), then any derailment that does occur needs to be treated in exactly the same way as it is on the real railway; an investigation must be carried out to establishe what happened and why to prevent a recurrence.  Once driver error is eliminated as a cause, the next question becomes 'has this happened at this spot before', in which case the primary supect is the track, followed by 'has this happened to this vehicle before', in which case the track is off the hook and the suspect is the vehicle.

 

If it's the track, check for level (transverse and longitudinal, especially on curves and gradients, and very especially where you have curves and gradients), doglegs, and crud in the flangeways.  I have a set of childrens' pound shop stiff nylon paint brushes for which I paid, well, you get the idea, which are carp for painting with but make first class crud brooms! 

 

If it's the stock, check the behaviour of the couplings (as has been said, tension locks are an imperfect way of transferring load between vehicles in a properly linear fashion and side forces are induced), the back to backs, the level of the axles in relation to the axleboxes and the rails, and the free running of the wheelsets, also the distribution of the weight of any load.  If it says 'Dapol' under the floor, replace the couplings and wheelsets anyway.  A good idea is to eliminate all, I mean all, plastic wheelsets from your layout; they run badly, are often off centre, cannot be adjusted if the B2B is out, and spread crud everywhere, and they look carp as well.

 

On bogie stock, check the above and that the bogies are free to swing vertically, transversly, and longitudinally, and that they and the wheelsets clear the solebars or anything else in the vicinity, and that the weight of the vehicle is evenly distributed between the boges along a longitudinal centre line.  Plastic pivot pins that engage in holes in the floor have a tendencey to foul on interior detail and to rub against the rim of the holes, so check this.  If lubrication of the interface is thought to be desireable, use a graphite stick or soft pencil.

 

Have a look at the buffers to ensure that the couplings are keeping them apart on your tightest curves, especially on long wheebase vehicles and bogie vehicles with long overhangs.  Try to establish standards for couplings and buffers in terms of height above the rails, and for tension locks, mountings, distance of the bar from the buffer beam, and bar/hook profile and material, as far as possible; complete standardisation is impossible because of the variety of situations in which these couplings must be mounted, but the manufacturers don't help with different bar and hook profiles and materials, just do the best you can!  Look at wagon weight and axle loading as well.  Much has been written about this, and a general consensus that 25g per 4 wheel wagon is about right, but it is in practice a moveable feast.  You will develop a feel for what is too heavy and what is too light, and particular issues may be encountered with unloaded Lowmacs and the like.

 

Finally, check the clearances for stock on adjacent lines, platforms, and scenery.  You may have passed these as fit for purpose when you built the layout and acquired locos or stock that are out of your particular loading gauge afterwards, and got away with it untl trains pass at a particular locaction and the stock happens to wobble sufficiently to contact, same with platforms and scenery.

 

A tip for clearances, which I think I might have read in a Model Railway Constructor article back in the mid 60s and which I think came from Cyril Freezer, no less, is the following, applied at the tracklaying stage,  With the track laid and pinned lightly in position so that further adjustments can be made before it is glued and ballasted, tape felt tip pens (in 1965 it was pencils) to the end corners and the central point of your longest vehicle, so that the pens are resting on the baseboard.  Then push the vehicle around the layout, everywhere, gently so that it is upright. 

 

You will now have lines, perhaps of different colours for ends and middles, alongside both sides of your newly pinned down track.  If any of these lines intersect, you have established a fouling point.  These are inevitable at turnouts and junctions, but need to be avoided between running lines and sidings.  You need to know the fouling points of your sidings so that you know how full they can safely be or how close a vehicle can be safely left at the turnout end.

 

You will have given yourself a safety margin to account for stock rocking in motion to the extent of half the diameter of the pens/pencils.  Repeat the excercise before the final postioning of the track is established.

[Il Grifone]

7 hours ago, RobinofLoxley said:

I recapped (repleaced all the capacitors) a 30 year NAD amplifier earlier this year. Sounded like new afterwards - 'sounded' in this case meaning absence of sound.

 

My NAD sounds fine though I suspect new electrolytics would still improve matters. Probably my Quad 33 would also be improved, though this suffers from crackles (apparently due to the plug-in modules, but hard to pin-point the exact source).

 

I have a boom-box type affair than hums even when switched off. Obviously a duff transformer (and cr poor design). It this case i think the fix involves a bin recycling centre.

[The Johnster]

Just checked my NADs and there are no capacitors.  There is capacity, but that's not quite the same thing...